[asterisk-bugs] [Asterisk 0016793]: T38 flow states HDLC carrier immediately down when writing TSI to fax, but were up

[Asterisk Bug Tracker]

A NOTE has been added to this issue. 
====================================================================== 
https://issues.asterisk.org/view.php?id=16793 
====================================================================== 
Reported By:                rlr2maverick
Assigned To:                
====================================================================== 
Project:                    Asterisk
Issue ID:                   16793
Category:                   Applications/app_fax
Reproducibility:            always
Severity:                   major
Priority:                   normal
Status:                     feedback
Asterisk Version:           1.6.1.14 
JIRA:                       SWP-890 
Regression:                 No 
Reviewboard Link:            
SVN Branch (only for SVN checkouts, not tarball releases): 1.6.1 
SVN Revision (number only!):  
Request Review:              
====================================================================== 
Date Submitted:             2010-02-09 09:10 CST
Last Modified:              2010-02-11 10:23 CST
====================================================================== 
Summary:                    T38 flow states HDLC carrier immediately down when
writing TSI to fax, but were up
Description: 
Good morning,

our setup is based on asterisk 1.6.1.14 & spandsp 0.0.6-pre17. We're
always receiving fax with no error.

When sending we're experiencing a quite strange chan_sip codecs
negotiation, selecting ulaw (8kbit??? with fax @14400 ???) even if slin is
preferred and available.

At the first write (TSI) from asterisk to the channel remote side (fax)
we're encountering HDLC carrier down and the transmission never takes
place.

T38 nego can't proceed and fax is never sent.

Is there any experience to share ?

Indeed,
RLR
====================================================================== 

---------------------------------------------------------------------- 
 (0117995) rlr2maverick (reporter) - 2010-02-11 10:23
 https://issues.asterisk.org/view.php?id=16793#c117995 
---------------------------------------------------------------------- 
dear lmadsen,

I'm considering a more specific case title as

"T.38 session fails with '488 Not acceptable here' if within 5 seconds
there is no no "SIP 100 Trying" REINVITE reply from remote"

The code aborting the T.38 sip session is in chan_sip.c

  19391                 case AST_STATE_UP:
  19392                         ast_debug(2, "%s: This call is UP.... \n",
c->name);
  19393 
  19394                         transmit_response(p, "100 Trying", req);
  19395 
  19396                         if (p->t38.state == T38_PEER_REINVITE) {
  19397                   --------------> p->t38id = ast_sched_add(sched,
5000, sip_t38_abort, dialog_ref(p, "passing dialog ptr into sched structure
based on t38id for sip_t38_abort."));

Nothing comes from the fax/ATA and the session is terminated (while
already sending out fax data).

Am I wrong when considering a patch where this delay is greater than 120
seconds  ? (I'll try) 

Issue History 
Date Modified    Username       Field                    Change               
====================================================================== 
2010-02-11 10:23 rlr2maverick   Note Added: 0117995                          
======================================================================